NVIDIA-AI-IOT / ros2_tao_pointpillars

ROS2 node for 3D object detection using TAO-PointPillars.
Apache License 2.0
70 stars 22 forks source link

Get individual point labels instead of bounding boxes #8

Closed tiesus closed 9 months ago

tiesus commented 9 months ago

Bounding boxes are quite strict. Is there an opportunity to get labels per point?